The goal of our proposal is to develop an innovative assay using reprogrammed, stem-like cells derived from primary pancreatic ductal adenocarcinomas (PDAs) to screen for new, personalized therapy. Our long- term goal is to discover effective anticancer drugs and cure PDA. Our application is highly relevant to the NIH Pancreatic Cancer Pilot Project initiative because we propose to: 1) Develop a novel cell-based assay with reprogrammed cells derived directly from patient tumors, and, 2) Harness the reprogrammed cells to screen for effective, personalized therapy with a unique library of clinic ready agents. Our innovative assay is based on the recent discovery that tumor and nonmalignant cells from patient biopsies can be expanded in culture by growing the cells under conditions similar to those used for embryonic stem cells. This strategy induces an adult stem cell phenotype in normal epithelial or carcinoma cells. Specifically, the combination of Rho kinase inhibitor (Y-27632) with mouse feeder cells reprograms epithelial cells into stem-like cells with the following properties: 1) indefinite self-renewal, 2) the ability to differentiate into mor mature cells from the tissue of origin when removed from the feeder cells and Y-27632. The cells (called conditionally reprogrammed cells or CRCs) derived from tumors faithfully replicate the original tumor histology in mouse xenografts, while CRCs from normal tissue are nontumorigenic. In a patient with refractory lung tumors (respiratory papillomatosis), CRCs were used to identify vorinostat as a drug that preferentially killed the tumor cells, but not normal tissue. This therapy subsequently led to a dramatic clinical response and reduction in tumor burden for several years. Moreover, the response of the tumor CRCs to drugs in vitro (including both sensitive and resistant agents) recapitulated the patient's responses to therapy in vivo. Based on these exciting results, we hypothesize that CRCs will be an innovative and useful model for PDA to uncover new, effective therapies. We propose to test this hypothesis with the following Specific Aims: 1) To generate CRCs from 20 patient tumors and matched control tissue, and, 2) To use a unique library of >3,000 drugs already approved for use in humans to screen for agents that selectively kill CRCs from PDA, but not normal tissue. This proposal is innovative because: 1) Until now, CRCs have not been generated from PDA samples and will provide a novel tool to study PDA, 2) CRCs have never been used to screen PDA for new therapeutic agents, 3) Because CRCs have properties of stem cells, they may serve as a model for cancer stem cells/tumor-initiator cells and help to identify therapy directed at this unique population of cells, 4) Co-culture with fibroblasts recapitulates the stromal compartment of the tumor microenvironment, 5) Our drug library is diverse, large, and agents are ready for use in patients. At the completion of this grant period, we expect to have a list of new drugs that are cytotoxic to PDA cells to test in preclinical studies and translate into clinical trials led by our clinical collaborators. This strategy could provide a new paradigm for personalized therapy for PDA.